These authors contributed equally to this paper.
Short Report: Complications
Magnetic resonance imaging retinal oximetry: a quantitative physiological biomarker for early diabetic retinopathy?
Article first published online: 12 MAR 2012
© 2011 The Authors. Diabetic Medicine © 2011 Diabetes UK
Volume 29, Issue 4, pages 501–505, April 2012
How to Cite
Yang, Y., Zhu, X. R., Xu, Q. G., Metcalfe, H., Wang, Z. C. and Yang, J. K. (2012), Magnetic resonance imaging retinal oximetry: a quantitative physiological biomarker for early diabetic retinopathy?. Diabetic Medicine, 29: 501–505. doi: 10.1111/j.1464-5491.2011.03440.x
- Issue published online: 12 MAR 2012
- Article first published online: 12 MAR 2012
- Accepted manuscript online: 13 SEP 2011 04:19PM EST
- Accepted 1 September 2011
- magnetic resonance imaging;
- retinal screening;
Diabet. Med. 29, 501–505 (2012)
Aims To assess the efficacy of using magnetic resonance imaging measurements of retinal oxygenation response to detect early diabetic retinopathy in patients with Type 2 diabetes.
Methods Magnetic resonance imaging was conducted during 100% oxygen inhalation in patients with Type 2 diabetes with either no diabetic retinopathy (n = 12) or mild to moderate background diabetic retinopathy (n = 12), as well as in healthy control subjects (n = 12). Meanwhile, changes in retinal oxygenation response were measured.
Results In the healthy control group, levels of retinal oxygenation response increased slowly during 100% oxygen inhalation. In contrast, they increased more quickly and attained homeostasis much earlier in the groups with background diabetic retinopathy (at the 20-min time point) and with no diabetic retinopathy (at the 25-min time point) than in the healthy control group (at the 42-min time point). Furthermore, levels of retinal oxygenation response in the group with background diabetic retinopathy increased more than that of the group with no diabetic retinopathy, which in turn increased more than that of the healthy control group. There are statistically significant differences between the group with background diabetic retinopathy and the healthy control group at 6-, 8-, 10-, 15-, 20- and 25-min time points (P < 0.05). According to the normal range of the healthy control group by setting fundus photography results as ‘gold standard’ in our research, the sensitivity, specificity, positive predictive value, negative predictive value and receiver operating characteristic area for reporting the early indications of utility of diabetic retinopathy were 83.33%, 58.33%, 50%, 87.5% and 0.774, respectively.
Conclusions The results indicate that magnetic resonance imaging is a potential screening method and probably a quantitative physiological biomarker to find early diabetic retinopathy in patients with Type 2 diabetes.